|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/istqb-ai-testing/
课程评论：没有评论
课程名称：ISTQB AI测试 - 学习最佳实践并为考试做准备 课程概述： 随着人工智能（AI）系统在我们日常生活中越来越普遍和重要，传统系统的测试方法已逐渐无法满足新出现的挑战。本课程将介绍人工智能的关键概念，如何制定接受标准以及如何测试基于AI的系统。这些系统具有独特的特征，如复杂性（例如深度神经网络）、自学习能力、基于大数据及非确定性，这给测试带来了新的挑战和机遇。 课程将介绍当今使用的多种类型的AI系统，并解释机器学习（ML）如何通常是这些系统的核心部分，同时展示构建ML系统的简易性。我们将探讨接受标准在AI系统中的变化需求，为什么要考虑伦理问题，以及AI系统特征使得测试比传统系统更具挑战性。 在ISTQB AI测试课程中，将从三个角度介绍如何实现质量。首先，考虑在构建机器学习系统时必须做出的选择和检查，确保用于训练和预测的数据质量。理想情况下，我们希望数据没有偏见和错误标记，并且与所要解决的问题紧密相关。接下来，将探讨适用于AI系统黑箱测试的多种方法，如回归测试和A/B测试，并详细介绍变态测试技术。第三，展示如何应用白箱测试以推动测试并衡量神经网络的测试覆盖率。通过自动驾驶汽车的案例，展示虚拟测试环境的必要性。 最后，将考虑利用AI作为支持测试工具的基础，展示AI成功应用于常见测试问题的实例。本课程具有很强的实用性，包含许多动手练习，为学员提供构建和测试多种类型机器学习系统的经验，无需编程背景。
Course OverviewThe testing of traditional systems is well-understood, but AI-based systems, which are becoming more prevalent and critical to our daily lives, introduce new challenges. This course will introduce the key concepts of Artificial Intelligence (AI), how we decide acceptance criteria and how we test AI-based systems. These systems have unique characteristics, which makes them special - they can be complex (e.g. deep neural nets), self-learning, based on big data, and non-deterministic, which creates many new challenges and opportunities for testing them.The course will introduce the range of types of AI-based systems in use today and explain how machine-learning (ML) is often a key part of these systems and show how easy it is to build ML systems. We will look at how the setting of acceptance criteria needs to change for AI-based systems, why we need to consider ethics, and show how the characteristics of AI-based systems make testing more difficult than for traditional systems.Introduction to ISTQB AI Testing Course by AITThree perspectives are used to show how quality can be achieved with these systems. First, we will consider the choices and checks that need to be made when building a machine-learning system to ensure the quality of data used for both training and prediction. Ideally, we want data that is free from bias and mis-labelling, but, most importantly, closely aligned with the problem. Next, we will consider the range of approaches suitable for the black-box testing of AI-based systems, such as back-to-back testing and A/B testing, introducing, in some detail, the metamorphic testing technique. Third, we will show how white-box testing can be applied to drive the testing and measure the test coverage of neural networks.The need for virtual test environments will be demonstrated using the case of self-driving cars as an example.Finally, the use of AI as the basis of tools that support testing will be considered by looking at examples of the successful application of AI to common testing problems.The course is highly practical and includes many hands-on exercises, providing attendees with experience of building and testing several different types of machine learning systems. No programming experience is required.
